David Fitzpatrick, University of Colorado, Boulder; Eric Sutton, Space Weather Technology, Research, and Education Center (SWx TREC), University of Colorado, Boulder; Marcin Pilinski, Laboratory for Atmospheric and Space Physics
Keywords: SpaceX/Starlink, Satellite Constellation, Thermosphere, Space Weather, Neutral Density
Abstract:
The growth in the satellite population in recent years brings to the forefront the challenge of quantifying thermospheric neutral density, a critical element of the LEO environment that prevails as the single largest source of error in the prediction of orbits. Unrealistic uncertainties in thermospheric neutral density have significant operational implications, posing challenges for satellite operators and the broader space industry in terms of collision risk assessment, station-keeping maneuver planning, and space traffic regulatory compliance. For many satellite operators, the unavailability of accurate on-orbit atmospheric density forecasts has also been shown to precipitate harsh financial consequences, as demonstrated by the February 2022 Starlink loss event, which is thought to have resulted in millions of dollars of loss. The promise of high-accuracy orbital prediction models is dependent upon being able to derive measurements of the upper atmosphere from widely available data products that scale with the spatial density of man-made objects in LEO.
In the broader context of this work, we seek to address this need by advancing the accuracy of “nowcasts” and by producing actionable forecasts of thermospheric conditions through the utilization of information from the decay of satellite orbits. To this end, SpaceX has partnered with NOAA’s National Environmental Satellite, Data, and Information Service (NESDIS) to provide us with continuous satellite ephemeris and housekeeping data from several thousand Starlink satellites. Unlike the small number of satellite missions carrying the scientific-grade accelerometers capable of producing in-situ measurements of neutral density along a single orbital path, the Starlink constellation has the distinct advantage of simultaneously covering vast swaths of the globe in the aggregate through its multiple orbital planes and string-of-pearls formations. However, compared to near-instantaneous accelerometer measurements, densities derived from the Starlink constellation must be averaged over some effective time (i.e., one full orbital period) to mitigate errors associated with the measured orbit ephemeris and satellite force model. Moreover, large uncertainties exist in the drag force model caused by a lack of knowledge of the underlying gas-surface collision physics and, to a lesser extent, the accuracy of a satellite’s geometry and pointing.
By continuously monitoring the dissipation of orbital energy with GNSS measurements, orbit-effective thermospheric density observations have been made for over 1,500 Starlink satellites. This effort is enabled by augmenting data from the Starlink onboard navigation filter with a high-fidelity, physics-based characterization of the nonconservative forces, including drag and lift forces, as well as solar radiation pressure. The improved force model is further refined through a least-squares minimization of data-to-model discrepancies between the densities derived from Starlink with those given by the High Accuracy Satellite Drag Model (HASDM), which is regarded by the U.S. Departments of Defense and Commerce as the state-of-the-art in thermospheric density models. To validate the accuracy of the developed force model, the resulting orbital densities are then compared to accelerometer-derived observations from the Gravity Recovery and Climate Experiment Follow-On (GRACE-FO) mission. The work then concludes with a commentary on the future phases of the research, including the proposal of a data-assimilative framework to ingest the derived densities into a physics-based, operational prediction model such as NOAA/Space Weather Prediction Center’s Whole Atmosphere Model (WAM). High-rate GNSS tracking data combined with comprehensive attitude and geometry knowledge, when made available by operators of large constellations such as Starlink, supports observations of the thermosphere with unprecedented coverage. Harnessing these observations is not without their challenges, and this work will elucidate the obstacles and constraints associated with utilizing such commercial datasets. Applying an innovative physics-based satellite force, the authors hope to contribute to a more comprehensive understanding of the thermospheric environment, ultimately enhancing space situational awareness and enhancing the sustainability of satellite operations in LEO.
Date of Conference: September 17-20, 2024
Track: Atmospherics/Space Weather